loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1
loaded library: loaded library: loaded library: loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1
/usr/lib/x86_64-linux-gnu/libibverbs.so.1/usr/lib/x86_64-linux-gnu/libibverbs.so.1/usr/lib/x86_64-linux-gnu/libibverbs.so.1

loaded library: 
/usr/lib/x86_64-linux-gnu/libibverbs.so.1
loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1
loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1
loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1
W20220705 13:56:55.858275 82934 rpc_client.cpp:190] LoadServer 127.0.0.1 Failed at 0 times error_code 14 error_message failed to connect to all addresses
W20220705 13:56:55.858311 82935 rpc_client.cpp:190] LoadServer 127.0.0.1 Failed at 0 times error_code 14 error_message failed to connect to all addresses
[07/05 13:57:07 libai]: Rank of current process: 0. World size: 8
[07/05 13:57:07 libai]: Command line arguments: Namespace(config_file='configs/bert_nl24_nah16_hs1024.py', eval_only=False, fast_dev_run=False, opts=['model.cfg.hidden_layers=24', 'train.dist.pipeline_num_layers=24', 'train.train_micro_batch_size=128', 'train.global_batch_size=2048', 'train.dist.tensor_parallel_size=1', 'train.dist.pipeline_parallel_size=4', 'train.amp.enabled=true', 'train.activation_checkpoint.enabled=true', 'train.train_iter=220', 'train.log_period=100', 'train.output_dir=test_logs/01b1d32/1n8g/LibAI_bert_nl24_nah16_hs1024_FP16_actrue_mp1_pp4_mb128_gb2048_1n8g_20220705_135654422062875'], resume=False)
[07/05 13:57:07 libai]: Contents of args.config_file=configs/bert_nl24_nah16_hs1024.py:
from libai.config import LazyCall
from libai.evaluation import PPLEvaluator
from .common.models.bert import pretrain_model as model
from .common.models.graph import graph
from .common.train import train
from .common.optim import optim
from .common.data.bert_dataset import dataloader, tokenization

#vocab_file = "/workspace/dataset/bert-base-chinese-vocab.txt"
#data_prefix = "/workspace/dataset/loss_compara_content_sentence"

vocab_file = "/dataset/source/dataset/bert-base-chinese-vocab.txt"
data_prefix = "/dataset/source/dataset/loss_compara_content_sentence"

tokenization.tokenizer.vocab_file = vocab_file
dataloader.train.dataset[0].data_prefix = data_prefix
dataloader.train.dataset[0].indexed_dataset.data_prefix = data_prefix
# dataloader.train.num_workers = 4

# Bert-large model config
#model.cfg.hidden_layers = 24
model.cfg.num_attention_heads = 16
model.cfg.hidden_size = 1024

#train.dist.pipeline_num_layers = model.cfg.hidden_layers
train.test_micro_batch_size = 4

train.evaluation.evaluator = LazyCall(PPLEvaluator)()

train.input_placement_device = "cpu"


train.evaluation.enabled = False
train.evaluation.eval_iter = 30

[07/05 13:57:07 libai]: Full config saved to test_logs/01b1d32/1n8g/LibAI_bert_nl24_nah16_hs1024_FP16_actrue_mp1_pp4_mb128_gb2048_1n8g_20220705_135654422062875/config.yaml
[07/05 13:57:07 lb.engine.default]: > compiling dataset index builder ...
make: Entering directory '/dataset/xyn/libai_bench/libai/libai/data/data_utils'
make: Nothing to be done for 'default'.
make: Leaving directory '/dataset/xyn/libai_bench/libai/libai/data/data_utils'
[07/05 13:57:07 lb.engine.default]: >>> done with dataset index builder. Compilation time: 0.040 seconds
[07/05 13:57:07 lb.engine.default]: >>> done with compiling. Compilation time: 0.041 seconds
[07/05 13:57:07 lb.engine.default]: Prepare training, validating, testing set
[07/05 13:57:07 lb.data.data_utils.indexed_dataset]: building dataset index ...
[07/05 13:57:07 lb.data.data_utils.indexed_dataset]: warming up index mmap file...
[07/05 13:57:07 lb.data.data_utils.indexed_dataset]: reading sizes...
[07/05 13:57:07 lb.data.data_utils.indexed_dataset]: reading pointers...
[07/05 13:57:07 lb.data.data_utils.indexed_dataset]: reading document index...
[07/05 13:57:07 lb.data.data_utils.indexed_dataset]: warming up data mmap file...
[07/05 13:57:07 lb.data.data_utils.indexed_dataset]: creating numpy buffer of mmap...
[07/05 13:57:07 lb.data.data_utils.indexed_dataset]: creating memory view of numpy buffer...
[07/05 13:57:07 lb.data.data_utils.indexed_dataset]: Finished creating indexed dataset in 0.100003 seconds
[07/05 13:57:07 lb.data.data_utils.indexed_dataset]: indexed dataset stats:
[07/05 13:57:07 lb.data.data_utils.indexed_dataset]: number of documents: 50000
[07/05 13:57:07 lb.data.data_utils.indexed_dataset]: number of sentences: 1249934
[07/05 13:57:07 lb.data.data_utils.dataset_utils]:  > loading indexed mapping from /dataset/source/dataset/loss_compara_content_sentence_bert_indexmap_450560mns_509msl_0.10ssp_1234s.npy
[07/05 13:57:07 lb.data.data_utils.dataset_utils]:     loaded indexed file in 0.005 seconds
[07/05 13:57:07 lb.data.data_utils.dataset_utils]:     total number of samples: 452417
[07/05 13:57:07 lb.data.data_utils.dataset_utils]:  > loading indexed mapping from /dataset/source/dataset/loss_compara_content_sentence_bert_indexmap_8mns_509msl_0.10ssp_1234s.npy
[07/05 13:57:07 lb.data.data_utils.dataset_utils]:     loaded indexed file in 0.001 seconds
[07/05 13:57:07 lb.data.data_utils.dataset_utils]:     total number of samples: 5884
[07/05 13:57:07 lb.data.data_utils.dataset_utils]:  > loading indexed mapping from /dataset/source/dataset/loss_compara_content_sentence_bert_indexmap_8mns_509msl_0.10ssp_1234s.npy
[07/05 13:57:07 lb.data.data_utils.dataset_utils]:     loaded indexed file in 0.001 seconds
[07/05 13:57:07 lb.data.data_utils.dataset_utils]:     total number of samples: 5884
[07/05 13:57:09 lb.engine.default]: Auto-scaling the config to train.train_iter=220, train.warmup_iter=0
[07/05 13:57:13 lb.engine.default]: Model:
BertForPreTraining(
  (bert): BertModel(
    (embeddings): BertEmbeddings(
      (vocab_embeddings): VocabEmbedding(num_embeddings=21248, embedding_dim=1024)
      (position_embeddings): Embedding(num_embeddings=512, embedding_dim=1024)
      (tokentype_embeddings): Embedding(num_embeddings=2, embedding_dim=1024)
      (embedding_dropout): Dropout(p=0.1, inplace=False)
    )
    (extended_attn_mask): BertExtendedAttnMask()
    (encoders): ModuleList(
      (0): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (1): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (2): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (3): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (4): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (5): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (6): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (7): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (8): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (9): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (10): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (11): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (12): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (13): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (14): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (15): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (16): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (17): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (18): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (19): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (20): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (21): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (22): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
      (23): TransformerLayer(
        (drop_path): Identity()
        (input_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (self_attention): MultiheadAttention(
          hidden_size=1024, num_heads=16, is_cross_attention=False
          (dropout): Dropout(p=0.1, inplace=False)
          (query_key_value): Linear1D(in_features=1024, out_features=3072, bias=True, parallel=col)
          (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=row)
        )
        (post_attention_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (mlp): MLP(
          bias_gelu_fusion=True, bias_dropout_fusion=True, dropout=0.1
          (dense_h_to_4h): Linear1D(in_features=1024, out_features=4096, bias=True, parallel=col)
          (dense_4h_to_h): Linear1D(in_features=4096, out_features=1024, bias=True, parallel=row)
        )
      )
    )
    (final_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
    (pooler): BertPooler(
      (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=col)
      (activation_func): Tanh()
    )
  )
  (cls_head): BertPreTrainingHeads(
    (predictions): BertLMPredictionHead(
      (dense): Linear1D(in_features=1024, out_features=1024, bias=True, parallel=data)
      (activation_func): GELU()
      (layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
    )
    (seq_relationship): Linear1D(in_features=1024, out_features=2, bias=True, parallel=data)
    (lm_logits): LMLogits()
    (loss_func): BertLoss(
      (lm_loss): ParallelCrossEntropyLoss()
    )
  )
)
WARNING [07/05 13:57:13 lb.scheduler.lr_scheduler]: warmup iters equals to zero, return CosineLR
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO Bootstrap : Using eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO NET/Plugin: Failed to find ncclNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO NET/Plugin: Failed to find ncclCollNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO Plugin Path : /opt/hpcx/nccl_rdma_sharp_plugin/lib/libnccl-net.so
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO P2P plugin IBext
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO NCCL_IB_PCI_RELAXED_ORDERING set by environment to 1.
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO NET/IB : Using [0]mlx5_1:1/RoCE ; OOB eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO Using network IBext
NCCL version 2.12.10+cuda11.2
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO Bootstrap : Using eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO NET/Plugin: Failed to find ncclNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO NET/Plugin: Failed to find ncclCollNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO Plugin Path : /opt/hpcx/nccl_rdma_sharp_plugin/lib/libnccl-net.so
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO P2P plugin IBext
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO NCCL_IB_PCI_RELAXED_ORDERING set by environment to 1.
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO NET/IB : Using [0]mlx5_1:1/RoCE ; OOB eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO Using network IBext
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO NCCL_IB_GID_INDEX set by environment to 3.
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO NCCL_IB_GID_INDEX set by environment to 3.
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO NCCL_IB_TIMEOUT set by environment to 23.
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO NCCL_IB_RETRY_CNT set by environment to 7.
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO NCCL_IB_TIMEOUT set by environment to 23.
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO NCCL_IB_RETRY_CNT set by environment to 7.
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO PXN Disabled as plugin is v4
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO Setting affinity for GPU 0 to 03ff,ffffffff
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO PXN Disabled as plugin is v4
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO Setting affinity for GPU 1 to 03ff,ffffffff
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO Trees [0] -1/-1/-1->1->0 [1] -1/-1/-1->1->0
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO Channel 00/02 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO Channel 01/02 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO Trees [0] 1/-1/-1->0->-1 [1] 1/-1/-1->0->-1
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO Channel 00 : 1[65020] -> 0[65010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO Channel 00 : 0[65010] -> 1[65020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO Channel 01 : 1[65020] -> 0[65010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO Channel 01 : 0[65010] -> 1[65020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82933:84094 [1] NCCL INFO comm 0x7f90d4e1ce90 rank 1 nranks 2 cudaDev 1 busId 65020 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO comm 0x7f1009c668e0 rank 0 nranks 2 cudaDev 0 busId 65010 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82932:84098 [0] NCCL INFO Launch mode Parallel
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Bootstrap : Using eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO NET/Plugin: Failed to find ncclNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO NET/Plugin: Failed to find ncclCollNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Plugin Path : /opt/hpcx/nccl_rdma_sharp_plugin/lib/libnccl-net.so
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO P2P plugin IBext
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO NCCL_IB_PCI_RELAXED_ORDERING set by environment to 1.
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO NET/IB : Using [0]mlx5_1:1/RoCE ; OOB eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Using network IBext
NCCL version 2.12.10+cuda11.2
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO Bootstrap : Using eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO NET/Plugin: Failed to find ncclNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO NET/Plugin: Failed to find ncclCollNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO Plugin Path : /opt/hpcx/nccl_rdma_sharp_plugin/lib/libnccl-net.so
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO P2P plugin IBext
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO NCCL_IB_PCI_RELAXED_ORDERING set by environment to 1.
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO NET/IB : Using [0]mlx5_1:1/RoCE ; OOB eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO Using network IBext
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO NCCL_IB_GID_INDEX set by environment to 3.
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO NCCL_IB_GID_INDEX set by environment to 3.
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO NCCL_IB_TIMEOUT set by environment to 23.
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO NCCL_IB_RETRY_CNT set by environment to 7.
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO NCCL_IB_TIMEOUT set by environment to 23.
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO NCCL_IB_RETRY_CNT set by environment to 7.
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO PXN Disabled as plugin is v4
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO Setting affinity for GPU 3 to 03ff,ffffffff
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO PXN Disabled as plugin is v4
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Setting affinity for GPU 2 to 03ff,ffffffff
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Channel 00/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO Trees [0] -1/-1/-1->1->0 [1] -1/-1/-1->1->0 [2] -1/-1/-1->1->0 [3] -1/-1/-1->1->0
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Channel 01/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Channel 02/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Channel 03/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Trees [0] 1/-1/-1->0->-1 [1] 1/-1/-1->0->-1 [2] 1/-1/-1->0->-1 [3] 1/-1/-1->0->-1
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO Channel 00 : 1[67020] -> 0[67010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Channel 00 : 0[67010] -> 1[67020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO Channel 01 : 1[67020] -> 0[67010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Channel 01 : 0[67010] -> 1[67020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO Channel 02 : 1[67020] -> 0[67010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Channel 02 : 0[67010] -> 1[67020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO Channel 03 : 1[67020] -> 0[67010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Channel 03 : 0[67010] -> 1[67020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO 4 coll channels, 4 p2p channels, 4 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO 4 coll channels, 4 p2p channels, 4 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO comm 0x7fb64b5e8130 rank 0 nranks 2 cudaDev 2 busId 67010 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82935:84093 [3] NCCL INFO comm 0x7f0994f4a510 rank 1 nranks 2 cudaDev 3 busId 67020 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82934:84091 [2] NCCL INFO Launch mode Parallel
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Bootstrap : Using eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO NET/Plugin: Failed to find ncclNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO NET/Plugin: Failed to find ncclCollNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Plugin Path : /opt/hpcx/nccl_rdma_sharp_plugin/lib/libnccl-net.so
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO P2P plugin IBext
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO NCCL_IB_PCI_RELAXED_ORDERING set by environment to 1.
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO NET/IB : Using [0]mlx5_1:1/RoCE ; OOB eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Using network IBext
NCCL version 2.12.10+cuda11.2
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO Bootstrap : Using eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO NET/Plugin: Failed to find ncclNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO NET/Plugin: Failed to find ncclCollNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO Plugin Path : /opt/hpcx/nccl_rdma_sharp_plugin/lib/libnccl-net.so
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO P2P plugin IBext
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO NCCL_IB_PCI_RELAXED_ORDERING set by environment to 1.
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO NET/IB : Using [0]mlx5_1:1/RoCE ; OOB eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO Using network IBext
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO NCCL_IB_GID_INDEX set by environment to 3.
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO NCCL_IB_GID_INDEX set by environment to 3.
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO NCCL_IB_TIMEOUT set by environment to 23.
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO NCCL_IB_RETRY_CNT set by environment to 7.
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO NCCL_IB_TIMEOUT set by environment to 23.
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO NCCL_IB_RETRY_CNT set by environment to 7.
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO PXN Disabled as plugin is v4
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Setting affinity for GPU 4 to 0fffff,fffffc00,00000000
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO PXN Disabled as plugin is v4
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO Setting affinity for GPU 5 to 0fffff,fffffc00,00000000
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Channel 00/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO Trees [0] -1/-1/-1->1->0 [1] -1/-1/-1->1->0 [2] -1/-1/-1->1->0 [3] -1/-1/-1->1->0
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Channel 01/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Channel 02/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Channel 03/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Trees [0] 1/-1/-1->0->-1 [1] 1/-1/-1->0->-1 [2] 1/-1/-1->0->-1 [3] 1/-1/-1->0->-1
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Channel 00 : 0[69010] -> 1[69020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO Channel 00 : 1[69020] -> 0[69010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Channel 01 : 0[69010] -> 1[69020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO Channel 01 : 1[69020] -> 0[69010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO Channel 02 : 1[69020] -> 0[69010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Channel 02 : 0[69010] -> 1[69020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO Channel 03 : 1[69020] -> 0[69010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Channel 03 : 0[69010] -> 1[69020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO 4 coll channels, 4 p2p channels, 4 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO 4 coll channels, 4 p2p channels, 4 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO comm 0x7f378d3363e0 rank 0 nranks 2 cudaDev 4 busId 69010 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82937:84099 [5] NCCL INFO comm 0x7f3c15182a60 rank 1 nranks 2 cudaDev 5 busId 69020 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82936:84095 [4] NCCL INFO Launch mode Parallel
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO Bootstrap : Using eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO NET/Plugin: Failed to find ncclNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO NET/Plugin: Failed to find ncclCollNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO Plugin Path : /opt/hpcx/nccl_rdma_sharp_plugin/lib/libnccl-net.so
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO P2P plugin IBext
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO NCCL_IB_PCI_RELAXED_ORDERING set by environment to 1.
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO NET/IB : Using [0]mlx5_1:1/RoCE ; OOB eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO Using network IBext
NCCL version 2.12.10+cuda11.2
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO Bootstrap : Using eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO NET/Plugin: Failed to find ncclNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO NET/Plugin: Failed to find ncclCollNetPlugin_v5 symbol.
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO Plugin Path : /opt/hpcx/nccl_rdma_sharp_plugin/lib/libnccl-net.so
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO P2P plugin IBext
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO NCCL_IB_PCI_RELAXED_ORDERING set by environment to 1.
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO NET/IB : Using [0]mlx5_1:1/RoCE ; OOB eth0:192.168.11.142<0>
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO Using network IBext
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO NCCL_IB_GID_INDEX set by environment to 3.
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO NCCL_IB_GID_INDEX set by environment to 3.
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO NCCL_IB_TIMEOUT set by environment to 23.
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO NCCL_IB_RETRY_CNT set by environment to 7.
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO NCCL_IB_TIMEOUT set by environment to 23.
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO NCCL_IB_RETRY_CNT set by environment to 7.
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO PXN Disabled as plugin is v4
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO Setting affinity for GPU 6 to 0fffff,fffffc00,00000000
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO PXN Disabled as plugin is v4
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO Setting affinity for GPU 7 to 0fffff,fffffc00,00000000
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO Channel 00/02 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO Trees [0] -1/-1/-1->1->0 [1] -1/-1/-1->1->0
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO Channel 01/02 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO Trees [0] 1/-1/-1->0->-1 [1] 1/-1/-1->0->-1
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO Channel 00 : 0[6b010] -> 1[6b020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO Channel 00 : 1[6b020] -> 0[6b010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO Channel 01 : 0[6b010] -> 1[6b020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO Channel 01 : 1[6b020] -> 0[6b010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO comm 0x7f533849b5f0 rank 0 nranks 2 cudaDev 6 busId 6b010 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82939:84097 [7] NCCL INFO comm 0x7fe76052cc10 rank 1 nranks 2 cudaDev 7 busId 6b020 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82938:84096 [6] NCCL INFO Launch mode Parallel
[07/05 13:57:22 lb.engine.trainer]: Starting training from iteration 0
[07/05 13:57:26 lb.models.utils.graph_base]: Start compling the train graph which may take some time. Please wait for a moment ...
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Setting affinity for GPU 6 to 0fffff,fffffc00,00000000
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO Setting affinity for GPU 7 to 0fffff,fffffc00,00000000
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO Trees [0] -1/-1/-1->1->0 [1] -1/-1/-1->1->0
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Channel 00/02 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Channel 01/02 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Trees [0] 1/-1/-1->0->-1 [1] 1/-1/-1->0->-1
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO Channel 00 : 1[6b020] -> 0[6b010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Channel 00 : 0[6b010] -> 1[6b020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO Channel 01 : 1[6b020] -> 0[6b010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Channel 01 : 0[6b010] -> 1[6b020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82933:82933 [1] NCCL INFO Setting affinity for GPU 1 to 03ff,ffffffff
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO Setting affinity for GPU 0 to 03ff,ffffffff
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO Channel 00/02 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82933:82933 [1] NCCL INFO Trees [0] -1/-1/-1->1->0 [1] -1/-1/-1->1->0
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO Channel 01/02 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO Trees [0] 1/-1/-1->0->-1 [1] 1/-1/-1->0->-1
iv-ybpu7pvmis5m57pm6ny1:82933:82933 [1] NCCL INFO Channel 00 : 1[65020] -> 0[65010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO Channel 00 : 0[65010] -> 1[65020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82933:82933 [1] NCCL INFO Channel 01 : 1[65020] -> 0[65010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO Channel 01 : 0[65010] -> 1[65020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO comm 0x557647a732a0 rank 1 nranks 2 cudaDev 7 busId 6b020 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO comm 0x5633ab4003f0 rank 0 nranks 2 cudaDev 6 busId 6b010 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82933:82933 [1] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82933:82933 [1] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82933:82933 [1] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82933:82933 [1] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82933:82933 [1] NCCL INFO comm 0x55857d24e120 rank 1 nranks 2 cudaDev 1 busId 65020 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO comm 0x556434a3bfe0 rank 0 nranks 2 cudaDev 0 busId 65010 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO Setting affinity for GPU 4 to 0fffff,fffffc00,00000000
iv-ybpu7pvmis5m57pm6ny1:82937:82937 [5] NCCL INFO Setting affinity for GPU 5 to 0fffff,fffffc00,00000000
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO Channel 00/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82937:82937 [5] NCCL INFO Trees [0] -1/-1/-1->1->0 [1] -1/-1/-1->1->0 [2] -1/-1/-1->1->0 [3] -1/-1/-1->1->0
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO Channel 01/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO Channel 02/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO Channel 03/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO Trees [0] 1/-1/-1->0->-1 [1] 1/-1/-1->0->-1 [2] 1/-1/-1->0->-1 [3] 1/-1/-1->0->-1
iv-ybpu7pvmis5m57pm6ny1:82935:82935 [3] NCCL INFO Setting affinity for GPU 3 to 03ff,ffffffff
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO Setting affinity for GPU 2 to 03ff,ffffffff
iv-ybpu7pvmis5m57pm6ny1:82935:82935 [3] NCCL INFO Trees [0] -1/-1/-1->1->0 [1] -1/-1/-1->1->0 [2] -1/-1/-1->1->0 [3] -1/-1/-1->1->0
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO Channel 00/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO Channel 01/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO Channel 02/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO Channel 03/04 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO Trees [0] 1/-1/-1->0->-1 [1] 1/-1/-1->0->-1 [2] 1/-1/-1->0->-1 [3] 1/-1/-1->0->-1
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Setting affinity for GPU 6 to 0fffff,fffffc00,00000000
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO Setting affinity for GPU 7 to 0fffff,fffffc00,00000000
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Channel 00/02 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO Trees [0] -1/-1/-1->1->0 [1] -1/-1/-1->1->0
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Channel 01/02 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Trees [0] 1/-1/-1->0->-1 [1] 1/-1/-1->0->-1
iv-ybpu7pvmis5m57pm6ny1:82937:82937 [5] NCCL INFO Channel 00 : 1[69020] -> 0[69010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO Channel 00 : 0[69010] -> 1[69020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82937:82937 [5] NCCL INFO Channel 01 : 1[69020] -> 0[69010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO Channel 01 : 0[69010] -> 1[69020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO Channel 00 : 1[6b020] -> 0[6b010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Channel 00 : 0[6b010] -> 1[6b020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82935:82935 [3] NCCL INFO Channel 00 : 1[67020] -> 0[67010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO Channel 00 : 0[67010] -> 1[67020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82937:82937 [5] NCCL INFO Channel 02 : 1[69020] -> 0[69010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO Channel 02 : 0[69010] -> 1[69020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Channel 01 : 0[6b010] -> 1[6b020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO Channel 01 : 1[6b020] -> 0[6b010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82935:82935 [3] NCCL INFO Channel 01 : 1[67020] -> 0[67010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO Channel 01 : 0[67010] -> 1[67020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82937:82937 [5] NCCL INFO Channel 03 : 1[69020] -> 0[69010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO Channel 03 : 0[69010] -> 1[69020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82935:82935 [3] NCCL INFO Channel 02 : 1[67020] -> 0[67010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO Channel 02 : 0[67010] -> 1[67020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82935:82935 [3] NCCL INFO Channel 03 : 1[67020] -> 0[67010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO Channel 03 : 0[67010] -> 1[67020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82937:82937 [5] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82937:82937 [5] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82937:82937 [5] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82937:82937 [5] NCCL INFO 4 coll channels, 4 p2p channels, 4 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO 4 coll channels, 4 p2p channels, 4 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82935:82935 [3] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82935:82935 [3] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82935:82935 [3] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82935:82935 [3] NCCL INFO 4 coll channels, 4 p2p channels, 4 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO 4 coll channels, 4 p2p channels, 4 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO comm 0x557647b0fb60 rank 1 nranks 2 cudaDev 7 busId 6b020 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO comm 0x5633ab49cb60 rank 0 nranks 2 cudaDev 6 busId 6b010 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO comm 0x561f2b748220 rank 0 nranks 2 cudaDev 4 busId 69010 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82937:82937 [5] NCCL INFO comm 0x558af89dab80 rank 1 nranks 2 cudaDev 5 busId 69020 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO comm 0x5572f9e4d650 rank 0 nranks 2 cudaDev 2 busId 67010 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82935:82935 [3] NCCL INFO comm 0x563e0a059d20 rank 1 nranks 2 cudaDev 3 busId 67020 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82938:84507 [6] NCCL INFO Launch mode Parallel
iv-ybpu7pvmis5m57pm6ny1:82932:84648 [0] NCCL INFO Setting affinity for GPU 0 to 03ff,ffffffff
iv-ybpu7pvmis5m57pm6ny1:82933:84554 [1] NCCL INFO Setting affinity for GPU 1 to 03ff,ffffffff
iv-ybpu7pvmis5m57pm6ny1:82932:84648 [0] NCCL INFO Channel 00/02 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82933:84554 [1] NCCL INFO Trees [0] -1/-1/-1->1->0 [1] -1/-1/-1->1->0
iv-ybpu7pvmis5m57pm6ny1:82932:84648 [0] NCCL INFO Channel 01/02 :    0   1
iv-ybpu7pvmis5m57pm6ny1:82932:84648 [0] NCCL INFO Trees [0] 1/-1/-1->0->-1 [1] 1/-1/-1->0->-1
iv-ybpu7pvmis5m57pm6ny1:82932:84648 [0] NCCL INFO Channel 00 : 0[65010] -> 1[65020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82933:84554 [1] NCCL INFO Channel 00 : 1[65020] -> 0[65010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82932:84648 [0] NCCL INFO Channel 01 : 0[65010] -> 1[65020] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82933:84554 [1] NCCL INFO Channel 01 : 1[65020] -> 0[65010] via P2P/IPC
iv-ybpu7pvmis5m57pm6ny1:82932:84648 [0] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82932:84648 [0] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82932:84648 [0] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82932:84648 [0] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82933:84554 [1] NCCL INFO Connected all rings
iv-ybpu7pvmis5m57pm6ny1:82933:84554 [1] NCCL INFO Connected all trees
iv-ybpu7pvmis5m57pm6ny1:82933:84554 [1] NCCL INFO threadThresholds 8/8/64 | 16/8/64 | 8/8/512
iv-ybpu7pvmis5m57pm6ny1:82933:84554 [1] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
iv-ybpu7pvmis5m57pm6ny1:82933:84554 [1] NCCL INFO comm 0x7f8bd095f7c0 rank 1 nranks 2 cudaDev 1 busId 65020 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82932:84648 [0] NCCL INFO comm 0x7f0ef4aa4240 rank 0 nranks 2 cudaDev 0 busId 65010 - Init COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82932:84648 [0] NCCL INFO Launch mode Parallel
iv-ybpu7pvmis5m57pm6ny1:82938:84509 [6] NCCL INFO Launch mode Parallel
iv-ybpu7pvmis5m57pm6ny1:82936:84682 [4] NCCL INFO Launch mode Parallel
iv-ybpu7pvmis5m57pm6ny1:82934:84676 [2] NCCL INFO Launch mode Parallel
iv-ybpu7pvmis5m57pm6ny1:82932:84628 [0] NCCL INFO Launch mode Parallel
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/07/05 14:14:14.365, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:14.365, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:14.369, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 55 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:14.370, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 55 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:14.372, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:14.373, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 11050 MiB, 21460 MiB
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/07/05 14:14:14.374, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 56 %, 32510 MiB, 11050 MiB, 21460 MiB
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/07/05 14:14:14.374, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 56 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:14.379, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.379, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:14.380, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.379, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:14.379, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:14.390, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 55 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.390, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 55 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:14.391, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 55 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.391, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 55 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:14.391, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 55 %, 32510 MiB, 11690 MiB, 20820 MiB
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/07/05 14:14:14.396, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 64 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:14.396, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:14.398, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 64 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:14.399, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:14.400, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:14.401, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:14.402, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:14.410, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 54 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:14.410, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 56 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:14.410, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 54 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:14.411, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 56 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:14.411, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 56 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:14.412, Tesla V100-SXM2-32GB, 470.57.02, 81 %, 53 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:14.413, Tesla V100-SXM2-32GB, 470.57.02, 81 %, 53 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:14.432, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.433, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.433, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.440, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 58 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:14.440, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 58 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:14.450, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 58 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.451, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 58 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.451, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 58 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.453, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 56 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:14.453, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 56 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:14.457, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:14.459, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:14.459, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:14.462, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.464, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.470, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 54 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:14.471, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 54 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:14.471, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 54 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:14.472, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 58 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.472, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 58 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:14.483, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:14.484, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:14.490, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 54 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:14.490, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 54 %, 32510 MiB, 14954 MiB, 17556 MiB
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/07/05 14:14:24.165, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 56 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:24.168, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 11690 MiB, 20820 MiB
2022/07/05 14:14:24.169, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:24.169, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 55 %, 32510 MiB, 11050 MiB, 21460 MiB
2022/07/05 14:14:24.170, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 56 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:24.170, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 54 %, 32510 MiB, 11022 MiB, 21488 MiB
2022/07/05 14:14:24.171, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 68 %, 32510 MiB, 14954 MiB, 17556 MiB
2022/07/05 14:14:24.172, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 80 %, 32510 MiB, 14954 MiB, 17556 MiB
[07/05 14:14:33 lb.utils.events]:  eta: 0:19:46  iteration: 99/220  consumed_samples: 204800  total_loss: 7.991  lm_loss: 7.294  sop_loss: 0.6979  time: 9.8802 s/iter  data_time: 0.0122 s/iter total_throughput: 207.28 samples/s lr: 5.82e-05  
[07/05 14:31:01 lb.utils.events]:  eta: 0:03:17  iteration: 199/220  consumed_samples: 409600  total_loss: 7.896  lm_loss: 7.202  sop_loss: 0.6945  time: 9.8792 s/iter  data_time: 0.0119 s/iter total_throughput: 207.30 samples/s lr: 3.21e-06  
[07/05 14:34:19 lb.utils.events]:  eta: 0:00:00  iteration: 219/220  consumed_samples: 450560  total_loss: 7.893  lm_loss: 7.199  sop_loss: 0.6939  time: 9.8789 s/iter  data_time: 0.1689 s/iter total_throughput: 207.31 samples/s lr: 1.01e-06  
[07/05 14:34:19 lb.engine.hooks]: Overall training speed: 218 iterations in 0:35:53 (9.8789 s / it)
[07/05 14:34:19 lb.engine.hooks]: Total training time: 0:35:53 (0:00:00 on hooks)
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO comm 0x7f533849b5f0 rank 0 nranks 2 cudaDev 6 busId 6b010 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO comm 0x7fe76052cc10 rank 1 nranks 2 cudaDev 7 busId 6b020 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO comm 0x5633ab4003f0 rank 0 nranks 2 cudaDev 6 busId 6b010 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO comm 0x557647a732a0 rank 1 nranks 2 cudaDev 7 busId 6b020 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82933:82933 [1] NCCL INFO comm 0x7f90d4e1ce90 rank 1 nranks 2 cudaDev 1 busId 65020 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO comm 0x7f1009c668e0 rank 0 nranks 2 cudaDev 0 busId 65010 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82938:82938 [6] NCCL INFO comm 0x5633ab49cb60 rank 0 nranks 2 cudaDev 6 busId 6b010 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82939:82939 [7] NCCL INFO comm 0x557647b0fb60 rank 1 nranks 2 cudaDev 7 busId 6b020 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82937:82937 [5] NCCL INFO comm 0x7f3c15182a60 rank 1 nranks 2 cudaDev 5 busId 69020 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO comm 0x7f378d3363e0 rank 0 nranks 2 cudaDev 4 busId 69010 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82933:82933 [1] NCCL INFO comm 0x55857d24e120 rank 1 nranks 2 cudaDev 1 busId 65020 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO comm 0x7fb64b5e8130 rank 0 nranks 2 cudaDev 2 busId 67010 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82935:82935 [3] NCCL INFO comm 0x7f0994f4a510 rank 1 nranks 2 cudaDev 3 busId 67020 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO comm 0x556434a3bfe0 rank 0 nranks 2 cudaDev 0 busId 65010 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82933:82933 [1] NCCL INFO comm 0x7f8bd095f7c0 rank 1 nranks 2 cudaDev 1 busId 65020 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82932:82932 [0] NCCL INFO comm 0x7f0ef4aa4240 rank 0 nranks 2 cudaDev 0 busId 65010 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82937:82937 [5] NCCL INFO comm 0x558af89dab80 rank 1 nranks 2 cudaDev 5 busId 69020 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82936:82936 [4] NCCL INFO comm 0x561f2b748220 rank 0 nranks 2 cudaDev 4 busId 69010 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82935:82935 [3] NCCL INFO comm 0x563e0a059d20 rank 1 nranks 2 cudaDev 3 busId 67020 - Destroy COMPLETE
iv-ybpu7pvmis5m57pm6ny1:82934:82934 [2] NCCL INFO comm 0x5572f9e4d650 rank 0 nranks 2 cudaDev 2 busId 67010 - Destroy COMPLETE
*****************************************
Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 
*****************************************